Recommender Documentation¶
Recommender¶
-
class
xframes.toolkit.recommend.
MatrixFactorizationModel
(model, ratings, user_col, item_col, rating_col)[source]¶ Recommender model.
-
classmethod
load
(path)[source]¶ Load a model that was saved previously.
Parameters: path : str
The path where the model files are stored. This is the same path that was passed to
save
. There are three files/directories based on this path, with extensions ‘.model’, ‘.ratings’, and ‘.metadata’.Returns: out : MatrixFactorizationModel
A model that can be used to predict ratings.
-
predict
(user, item)[source]¶ Predict the rating given by a user to an item.
Parameters: user : int
The user to predict.
item : int
The item to rate.
Returns: out : float
The predicted rating.
-
predict_all
(user)[source]¶ Predict ratings for all items.
Parameters: user : int
The user to make predictions for.
Returns: out : XFrame
Each row of the frame consists of a user id, an item id, and a predicted rating.
-
recommend_top_k
(user, k=10)[source]¶ Recommend some items for a user.
Parameters: user : int
The user to make recommendations for.
Returns: out : XFrame
A XFrame containing the highest predictions for the user. The items that the user has explicitly rated are excluded.
-
save
(path)[source]¶ Save a model.
The model can be saved, then reloaded later to provide recommendations.
Parameters: path : str
The path where the model will be saved. This should refer to a file, not to a directory. Three items will be stored here: the underlying model parameters, the original ratings, and the column names. These are stored with suffix ‘.model’, ‘.ratings’, and ‘.metadata’.
-
classmethod
-
xframes.toolkit.recommend.
create
(data, user_col, item_col, rating_col, recommender_type='ALS', rank=50, iterations=10, lambda_=0.01, seed=0, **kwargs)[source]¶ Create a recommendation model.
Parameters: data : XFrame
A table containing the user ratings. This table must contin three columns corresponding to the users, the items, and the ratings. The table may contain other columns as well: these are not used.
user_col : string
The column name of the users.
item_col : string
The column name of the items.
rating_col : string
The column name of the ratings. This must be a number.
recommender_type : string, optional
The type of recommender. Optons are: * ALS * ALS-implicit
rank : int, optional
See
ALSBuilder.train
iterations : int, optional
See
ALSBuilder.train
lambda_ : float, optional
See
ALSBuilder.train
other : various, optional
See optional arguments to pyspark.mllib.recommendations.train.